Role of Bacterial and Host DNases on Host-Pathogen Interaction during Streptococcus suis Meningitis

Abstract

Streptococcus suis is a zoonotic agent causing meningitis in pigs and humans. Neutrophils, as the first line of defense against S. suis infections, release neutrophil extracellular traps (NETs) to entrap pathogens. In this study, we investigated the role of the secreted nuclease A of S. suis (SsnA) as a NET-evasion factor in vivo and in vitro. Piglets were intranasally infected with S. suis strain 10 or an isogenic ssnA mutant. DNase and NET-formation were analyzed in cerebrospinal fluid (CSF) and brain tissue. Animals infected with S. suis strain 10 or S. suis 10ΔssnA showed the presence of NETs in CSF and developed similar clinical signs. Therefore, SsnA does not seem to be a crucial virulence factor that contributes to the development of meningitis in pigs. Importantly, DNase activity was detectable in the CSF of both infection groups, indicating that host nucleases, in contrast to bacterial nuclease SsnA, may play a major role during the onset of meningitis. The effect of DNase 1 on neutrophil functions was further analyzed in a 3D-cell culture model of the porcine blood-CSF barrier. We found that DNase 1 partially contributes to enhanced killing of S. suis by neutrophils, especially when plasma is present. In summary, host nucleases may partially contribute to efficient innate immune response in the CSF.

Keywords: DNase; NETs; Streptococcus suis; meningitis; neutrophils; pathogenesis.

Figure 1

The DNase mutant S. suis10ΔssnA suggest only a slight attenuation in pigs. The body temperature and survival rate after intranasal infection with S. suis is presented: (A) The body temperature increased in both infection groups post-infection. Data shown as mean ± SD. Each dot represents one animal, red marks show animals with meningitis. Statistical analysis: one-tailed, paired Student’s t-test was calculated in each infection group. Values post-infection present the highest measured value. *** p < 0.001. (B) The Kaplan–Meier survival curve shows a slightly higher but not significant mortality in the S. suis 10-infected group (Statistical analysis: Log-rank (Mantel–Cox) test, p = 0.26, each group n = 9).

Figure 2

Neutrophil extracellular traps (NETs) are formed in the cerebrospinal fluid (CSF) of piglets infected with S. suis (strain 10) or its isogenic ssnA mutant and contain PR-39. Cytospins were conducted with CSF taken from infected piglets in the early phase experiment and were analyzed with immunofluorescence staining. (A) NET fibers were seen in CSF of piglets with meningitis from both infection groups (blue = DNA (Hoechst); green = DNA/histone-1-complexes (NETs)). Representative pictures are shown. (B) The antimicrobial peptide PR-39 is embedded in NET fibers and around the nucleus of neutrophils in CSF of infected piglets with meningitis from both infection groups. DNA/histone-1-complexes appeared to a lesser extent. The 3D image was modeled out of 50 z-stacks (0.17 µm steps) with LAS X 3D Version 3.1.0 software from Leica. (blue = DNA (Hoechst); green = DNA/histone-1-complexes (NETs); red = PR-39). Representative pictures are shown (scale bar A = 20 µm; B = 5 µm). (C) A high amount of antimicrobial peptide PR-39 in CSF of animals with clinical meningitis (red marks) was determined by ELISA. Data shown as mean ± SD. Statistical analysis: unpaired Mann–Whitney test.

Figure 3

NET-markers are present in brain tissue of piglets with meningitis in situ, but no NET-fibers are detectable. (A) In the staining of formalin-fixed brain tissue from pigs out of survival experiments, NET markers and S. suis were visualized, but no NET fibers. * cortex cerebri and # inflammation in meninges (blue = counterstaining of DNA (DAPI); green = DNA/histone-1-complexes (NETs); red = S. suis (white arrow); iso = isotype control; 4009 = animal with meningitis from this survival experiment study; 9899, 9884 and 9807 = pigs with meningitis from other survival experiment study [2,14] (scale bar upper panel = 50 µm; lower panel = 20 µm). (B) Representative hematoxylin-eosin (HE) stained tissues of meningitis show neutrophils invading from vessels into the meninges. This picture was observed in both infection groups of the early phase experiment. NET markers (green = DNA/histone-1-complexes (NETs) and yellow = neutrophil elastase) and Streptococci (red) were found in brain tissue of these animals in both infection groups; however, as in the survival experiment, no NET fibers were detectable. Representative pictures are shown; white arrows = S. suis. The pictures for both infection groups present slides that were sliced successively (scale bar HE, isotype and S. suis 10 = 50 µm, S. suis 10ΔssnA = 20 µm). Respective isotype controls were used to adjust the settings.

Figure 4

Host DNases are detectable and active during S. suis meningitis in choroid plexus, CSF and serum. (A) By immunofluorescence staining, we could show that the amount of DNase 1 in the choroid plexus is different in animals with inflammation, compared to animals without inflammation of the choroid plexus. In the choroid plexus of an infected pig without clinical signs of CNS disorders only weak DNase 1 signals were detected in the early phase experiment. In pigs with clinical signs of CNS disorders a high DNase 1 signal in the region of the choroid plexus epithelial cells was observed. PR-39 signal, as a marker for neutrophils, was found in these pigs as well (blue = DNA (Hoechst); green = PR-39; red = DNase 1 (scale bar upper panel = 50 µm; lower panel = 20 µm; arrows mark transmigrating or already transmigrated PR-39 positive neutrophils). Activity of DNases was observed in serum (B) and in CSF (C) of all pigs of the early phase experiment. A significant increase of DNase activity post-infection was found in serum for both infection groups after 48 h and for the S. suis 10-infected group also after 96 h. In CSF after 96 h, a higher increase was observed than after 48 h in both infection groups. Data shown as mean ± SD. Statistical analysis: unpaired Mann–Whitney test (* p < 0.05, ** p < 0.01).

Figure 5

DNase 1 has no impact on transmigration of isolated porcine neutrophils through a cell layer of porcine choroid plexus epithelial cells and killing of S. suis by neutrophils. (A) An inverted cell culture model with porcine choroid plexus epithelial (PCP-R) cells was used to investigate transmigration of neutrophils after infection with S. suis. PCP-R cells were grown on the underside of a filter insert with 3 µm pores. S. suis 10 transmigrated from the blood compartment at the basolateral side into the medium-filled CSF compartment at the apical side of PCP-R cells. The medium with bacteria in the blood compartment was removed and replaced by medium containing freshly isolated porcine neutrophils (polymorph nuclear cells (PMN)). Neutrophils were allowed to transmigrate through the cell layer for 4 h and counteract against S. suis in the CSF compartment. As read-out neutrophil/mL and CFU/mL were determined. (B) Stimulation with 100 ng/mL interleukin 8 (IL8) leads to high transmigration of neutrophils into the CSF compartment in the cell culture system. (C) The CFU/mL was not significantly influenced by transmigrated neutrophils with or without DNase 1 treatment. (+) component present in the well, (−) component not present in the well. Data shown as mean ± SD; n = 3 independent experiments in duplicates. Statistical analysis: one-way ANOVA followed by Kruskal–Wallis multiple comparison test (* p < 0.05, ** p < 0.01).

Figure 6

Cell culture model of the porcine blood–CSF barrier mimicking physiological conditions to investigate infection and transmigration. Neutrophils transmigrate out of fresh heparinized porcine whole blood through a layer of porcine choroid plexus epithelial cells (PCP-R) into porcine CSF. The CSF was enriched with interleukin 8 (IL8) to attract neutrophils. In addition, combinations with S. suis 10 infections and DNase 1 treatments were used. After 4 h of incubation, CFU/mL was determined by plating serial dilutions on blood agar plates and the number of transmigrated cells was determined by flow cytometry. Immunofluorescence staining of transmigrated neutrophils was conducted to analyze NET formation.

Figure 7

DNase 1 has no impact on transmigration of neutrophils trough a cell layer of PCP-R and killing of S. suis is donor-dependent. (A) NET formation was detectable in the CSF compartment in the porcine blood–CSF barrier mimicking physiological conditions in the absence of DNase 1, but was completely destroyed by DNase 1 (blue = DNA (Hoechst); green = DNA/histone-1-complexes (NETs); red = elastase). Representative pictures are shown (scale bar = 20 µm). (B) Transmigration of neutrophils out of fresh porcine blood trough PCP-R was neither significantly influenced by S. suis 10 nor DNase 1. Data shown as mean ± SD, n = 6 independent experiments with five duplicates and once a single value. Statistical analysis one-way ANOVA. (C) Killing of S. suis 10 in CSF was not significantly influenced by transmigrated neutrophils in presence of DNase 1. Statistical analysis: unpaired Student’s t-test. Data shown as mean ± SD, n = 6 independent experiments with five duplicates and once a single value. (D) Values from one experiment, with blood from the same donor pig, without and with DNase 1 in the CSF compartment are connected by a line. A donor-dependent influence regarding the effect of DNase 1 on killing efficiency is detectable (orange = DNase 1 reduces CFU; blue = DNase 1 increases CFU; violet = DNase 1 no impact on CFU). (E) Correlation of transmigrated neutrophils and S. suis 10 in CSF. The more neutrophils transmigrate into the CSF, the higher is the killing of the bacteria. Statistical analysis: Pearson correlation r = −0.5. (F) Correlation of transmigrated neutrophils and S. suis 10 + DNase 1 in CSF. With DNase 1 the Pearson correlation r = −0.8 is significant, p = 0.0026. E and F are calculated based on data from B and C.

Figure 8

DNase 1 improves killing of S. suis by neutrophils in the presence of plasma. (A) Immunofluorescence staining of NETs and intra- and extracellular S. suis. NET production and phagocytosis exist in parallel during neutrophil killing assay. With plasma, less bacteria can be seen than without plasma. More Streptococci are extracellular than intracellular as the single channels visualize (blue = DNA (Hoechst); magenta = DNA/histone1-complexes (NETs, white arrowheads); red and green = extracellular S. suis; green = intracellular S. suis (white arrows)). Representative pictures are shown (scale bar = 50 µm, zoom = 10 µm). (B) The effect of DNase 1 on killing of S. suis 10cpsΔEF was proven with a neutrophil killing assay with or without 10% plasma of the blood-donating pig. After 2 h incubation with plasma, an increased killing of S. suis was detectable with DNase 1. Without plasma, the CFU/mL was significantly higher than with plasma. The best killing effect was detectable with plasma and DNase 1. Data shown as mean ± SD, n = 5 independent experiments. Statistical analysis: paired students t-test (* p < 0.05, ** p < 0.01).